Install Tesseract OCR 5 on Rocky Linux 8|AlmaLinux 8

In day-to-day activities, we encounter situations where we need to extract text from documents. This is where OCR software comes in. OCR (Optical Character Recognition) is a technology used to recognize the text in images or scanned documents. Tesseract is one of the software in this category.

Tesseract is a free and open-source OCR originally developed by Hewlett-Packard. I was later open-sourced by HP in 2005 and developed by Google since 2006. The current stable version of Tesseract 5.-0 was released on November 30, 2021. This release has several improvements and bug fixes. Some of the features of Tesseract OCR 5 include:

• Added support for Apple Silicon
• Low memory consumption as compared to earlier versions
• Support for using floats for LSTM model training and text recognition.

This guide offers a deep illustration of how to install Tesseract OCR 5 on Rocky Linux 8|AlmaLinux 8.

Let’s dive in!

Install Tesseract OCR 5 on Rocky Linux 8|AlmaLinux 8

There are a couple of methods one can use to install Tesseract OCR 5 on Rocky Linux 8|AlmaLinux 8. In this guide, we will cover the two methods below:

• Building from Source
• Using Docker/Podman Containers

Method 1 – Install Tesseract OCR 5 on Rocky Linux 8|AlmaLinux 8 from Source

In this step, we will start by installing the required packages to build Tesseract OCR 5 from a source file.

sudo yum install git automake make autoconf libtool clang gcc-c++.x86_64

There are more leptonica dependencies required.

sudo yum install zlib zlib-devel libjpeg libjpeg-devel libwebp libwebp-devel libtiff libtiff-devel libpng libpng-devel

Move the executables to your path.

cd /usr/local/lib
sudo cp /usr/lib64/libjpeg.so.62 .
sudo cp /usr/lib64/libwebp.so.7 .
sudo cp /usr/lib64/libtiff.so.5 .
sudo cp /usr/lib64/libpng16.so.16 .

Compile Leptonica from source. Begin by cloning it from git as below.

cd ~
git clone https://github.com/DanBloomberg/leptonica.git --depth 1
cd leptonica

Now compile and install Leptonica.

./autogen.sh
./configure --prefix=/usr/local --disable-shared --enable-static --with-zlib --with-jpeg --with-libwebp  --with-libtiff --with-libpng --disable-dependency-tracking
make
sudo make install
sudo ldconfig

Once Leptonica is installed, we will proceed and download Tesseract OCR 5 from Github. You can also pull the file using Wget:

cd ~
VER=$(curl -s https://api.github.com/repos/tesseract-ocr/tesseract/releases/latest|grep tag_name | cut -d '"' -f 4)
wget https://github.com/tesseract-ocr/tesseract/archive/refs/tags/$VER.tar.gz -O tesseract-5.tar.gz

Extract the archive.

tar zxvf tesseract-5.tar.gz

Navigate into the extracted directory.

cd tesseract-*/

Now compile Tesseract OCR 5.

export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig
./autogen.sh
./configure --prefix=/usr/local --disable-shared --enable-static --with-extra-libraries=/usr/local/lib/ --with-extra-includes=/usr/local/lib/
make

Install Tesseract OCR 5 using the command:

sudo make install
sudo ldconfig

Load tesseract languages

Load tesseract languages by creating a language path.

mkdir -p /tess/traineddata

Export the Tesseract path by adding the below line to ~/.bashrc.

export TESSDATA_PREFIX=/home/$USER/tess/traineddata

You can replace $USER with the exact username on the system

Source the profile.

source ~/.bashrc

Now add any trained data available on Github tessdata to the path.

cd $TESSDATA_PREFIX
wget https://github.com/tesseract-ocr/tessdata/raw/main/eng.traineddata
wget https://github.com/tesseract-ocr/tessdata/raw/main/fra.traineddata

Method 2 – Install Tesseract OCR 5 on Rocky Linux 8|AlmaLinux 8 Using Docker/Podman

I assume that you already have Docker/Podman Engine installed on your system. Otherwise, install Docker Engine on Rocky Linux 8|AlmaLinux 8 using the guide below:

• How To Install Docker CE on Linux Systems

Podman can be installed using the command below:

sudo yum install podman

Once the preferred container engine has been installed, create the Tesseract OCR 5 docker file.

vim Dockerfile

In the file, add the lines below

FROM ubuntu
RUN apt-get update -y
ENV DEBIAN_FRONTEND=noninteractive 
RUN apt-get install -y gnupg apt-transport-https apt-utils wget
RUN echo "deb https://notesalexp.org/tesseract-ocr5/focal/ focal main" \
|tee /etc/apt/sources.list.d/notesalexp.list > /dev/null
RUN wget -O - https://notesalexp.org/debian/alexp_key.asc | apt-key add -
RUN apt-get update -y
RUN apt-get install tesseract-ocr -y
RUN apt install imagemagick -y
ENTRYPOINT ["tesseract"]
RUN tesseract -v

Save the file and build the Tesseract 5 image:

$ docker build . -t tesseract:5
##OR
$ podman build . -t tesseract:5

Once build you will see this

Removing intermediate container a9ecd5a7810f
 ---> e1f76f250bc1
Step 10/11 : ENTRYPOINT ["tesseract"]
 ---> Running in 1886dfbcfd5e
Removing intermediate container 1886dfbcfd5e
 ---> 30afc1531eb9
Step 11/11 : RUN tesseract -v
 ---> Running in 598972cbd362
tesseract 5.0.1
 leptonica-1.79.0
  libgif 5.1.4 : libjpeg 8d (libjpeg-turbo 2.0.3) : libpng 1.6.37 : libtiff 4.1.0 : zlib 1.2.11 : libwebp 0.6.1 : libopenjp2 2.3.1
 Found AVX2
 Found AVX
 Found FMA
 Found SSE4.1
 Found OpenMP 201511
 Found libarchive 3.4.0 zlib/1.2.11 liblzma/5.2.4 bz2lib/1.0.8 liblz4/1.9.2 libzstd/1.4.4
 Found libcurl/7.68.0 OpenSSL/1.1.1f zlib/1.2.11 brotli/1.0.7 libidn2/2.2.0 libpsl/0.21.0 (+libidn2/2.2.0) libssh/0.9.3/openssl/zlib nghttp2/1.40.0 librtmp/2.3
Removing intermediate container 598972cbd362
 ---> 35f8f89ea444
Successfully built 35f8f89ea444
Successfully tagged tesseract:5

Check the image:

##For Docker 
$ docker images
REPOSITORY   TAG       IMAGE ID       CREATED          SIZE
tesseract    5         35f8f89ea444   14 seconds ago   302MB

##For Podman
$ podman images
REPOSITORY                        TAG         IMAGE ID      CREATED        SIZE
localhost/tesseract               5           f92efe531f58  2 minutes ago  309 MB

With the Tesseract image, you can now run Tesseract using the syntax.

##For Docker
docker run --rm -v /path/to/image/:/tmp/:z tesseract:5 /tmp/img.jpg stdout

##For Podman
podman run --rm -v /path/to/image/:/tmp/:z localhost/tesseract:5 /tmp/img.jpg stdout

In the above command:

• –rm removes the conatiner after running the above command
• –v specifies the volume with /path/to/image/ as the local path to be mounted on /tmp/ of the conatiner and /tmp/img.jpg is the exact location of the image file the mounted local path.
• tesseract:5 is the conatiner image created
• z modifies selinux label. It indicates that the bind mount content is shared among multiple containers

Use Tesseract OCR 5 on Rocky Linux 8|AlmaLinux 8.

Once Tesseract OCR 5 has been installed, you can now start extracting text from scanned documents or images.

To convert an image to a text file, the syntax used is as below.

tesseract <image_name> <output file_name>

For example, converting an image to a text file the command will be:

tesseract image.png new

The output will be a text file- new of the image file- image.png.

For Docker/Podman, the command will be as below assuming the image file image.png is at /home/thor/Desktop/image.png

##For Docker
docker run --rm -v /home/thor/Desktop/:/tmp/:z tesseract:5 /tmp/image.png stdout

##For Podman
podman run --rm -v /home/thor/Desktop/:/tmp/:z localhost/tesseract:5 /tmp/image.png stdout

Or alternatively, shorten the command as below(Example with Podman) after navigating into the directory with the image file.

podman run --rm -v $(pwd)/:/tmp/:z localhost/tesseract:5 /tmp/image.png stdout 

This is the text output of the image.png in the directory

You can also have the output above to a file say new.txt as below.

tesseract image.png new

Example for Docker:

docker run --rm -v $(pwd)/:/tmp/:z tesseract:5 /tmp/image.png /tmp/new

Or to a PDF file(new.pdf) as below.

tesseract image.png new pdf

Example for docker

docker run --rm -v $(pwd)/:/tmp/:z tesseract:5 /tmp/image.png /tmp/new pdf

Specify the language.

When using Tesseract OCR you can specify the language you want to use with the -l flag. For example, use Czech.

tesseract image.png new -l ces

You can specify multiple languages as well.

tesseract image.png new -l ces+eng

The end!

Closing Thoughts.

We have triumphantly installed and used Tesseract OCR 5 on Rocky Linux 8|AlmaLinux 8. You can now extract text from images and scanned documents easily. I hope this was helpful.

See more: